AIAA Paper 2009-94. Presented at the 47th AIAA Aerospace Sciences Meeting Including The New Horizons Forum and Aerospace Exposition, January 5, 2009, pages 1-17. This article is in the public domain. Published by American Institute of Aeronautics and Astronautics.
NOTE: At the time of publication, the author Russell M. Cummings was on sabbatical leave from Cal Poly.
Unforeseen nonlinear aerodynamic behavior and/or fluid-structure interactions have affected the development of nearly every major fighter program since 1960. The development cost of each of these aircraft could have been drastically reduced if these aerodynamic issues had been identified earlier in the design process. Therefore, a high-fidelity computational tool capable of reliably predicting or identifying configurations susceptible to handling quality instabilities prior to flight test would be of great interest to the stability and control community. The United States Air Force Academy Modeling and Simulation Research Center and the United States Air Force Seek Eagle Office have initiated a joint effort to develop nonlinear lower-order aerodynamic loads models from unsteady CFD solutions. A key step in the process is to perform “training maneuvers,” which are dynamic CFD simulations designed to excite the relevant nonlinear flow physics. A reduced-order model is then built using SIDPAC, a regression based modeling technique designed specifically for aircraft system identification. The approach is validated for an aircraft configuration with a known aerodynamic instability that occurs well within the flight envelope. The dynamic CFD simulations can reliably predict this instability for frequencies ranging from 1.43 to 17.1 Hertz. In addition, an aerodynamic model trained using a varying frequency chirp maneuver was then used to predict constant frequency aerodynamic loads at conditions where strongly nonlinear aerodynamic behavior occurred.